Preparing the Oracle Solaris Cluster Nodes

Preparing the Oracle Solaris Cluster nodes modifies the configuration of the operating system to enable Oracle RAC to run on Oracle Solaris Cluster nodes. Preparing the Oracle Solaris Cluster nodes and disks involves the following tasks:

Bypassing the NIS name service
Creating the database administrator (DBA) group and the DBA user accounts
Configuring shared memory for the Oracle RAC software

Caution - Perform these tasks on all nodes where Support for Oracle RAC can run. If you do not perform these tasks on all nodes, the Oracle installation is incomplete. An incomplete Oracle installation causes Support for Oracle RAC to fail during startup.

To enable the Oracle RAC to run in a zone cluster, you need to perform the following additional tasks:

Configuring shared memory for Oracle RAC software in a zone cluster
Setting necessary privileges for Oracle RAC software in a zone cluster
Configuring logical hostname resources for Oracle RAC software in a zone cluster

This section contains the following information:

Before You Begin
How to Bypass the NIS Name Service
How to Create the DBA Group and the DBA User Accounts
How to Configure Shared Memory for Oracle RAC Software in the Global Cluster
How to Configure Shared Memory for Oracle RAC Software in a Zone Cluster
How to Set the Necessary Privileges for Oracle RAC Software in a Zone Cluster
How to Configure the Logical Hostname Resources or Virtual IP Addresses for Oracle RAC Software in a Zone Cluster

Before You Begin

Before you prepare the Oracle Solaris Cluster nodes, ensure that all preinstallation tasks for Oracle RAC are completed. For more information, see your Oracle RAC documentation.

How to Bypass the NIS Name Service

Bypassing the NIS name service protects the Support for Oracle RAC data service against a failure of a cluster node's public network. A failure of a cluster node's public network might cause the NIS name service to become unavailable. If Support for Oracle RAC refers to the NIS name service, unavailability of the name service might cause the Support for Oracle RAC data service to fail.

Bypassing the NIS name service ensures that the Support for Oracle RAC data service does not refer to the NIS name service when the data service sets the user identifier (ID). The Support for Oracle RAC data service sets the user ID when the data service starts or stops the database.

Become superuser on all nodes where Support for Oracle RAC can run.

On each node, include the following entries in the /etc/nsswitch.conf file.

passwd:    files [NOTFOUND=return] nis [TRYAGAIN=0]
publickey: files [NOTFOUND=return] nis [TRYAGAIN=0]
project:   files [NOTFOUND=return] nis [TRYAGAIN=0]
group:     files [NOTFOUND=return] nis [TRYAGAIN=0]

For more information about the /etc/nsswitch.conf file, see the nsswitch.conf(4) man page.

Next Steps

Go to How to Create the DBA Group and the DBA User Accounts.

How to Create the DBA Group and the DBA User Accounts

In installations of Oracle RAC with Oracle Solaris Cluster software, the DBA group is normally named dba. This group normally contains the root user and the oracle user.

Note - This configuration of users and groups differs from the configuration that is described in the Oracle documentation for a standalone installation of Oracle RAC. A standalone installation of Oracle RAC uses a primary DBA group that is named oinstall and a secondary group that is named dba. Some applications also require a secondary group that is named oper. For more information, see your Oracle documentation.

Perform this task on each cluster node.

On the cluster node where you are performing this task, become superuser.
Add an entry for the DBA group and potential users in the group to the /etc/group file.
```
# groupadd -g group-id group-name
```
group-name

Specifies the name of the group for which you are adding an entry. This group is normally named dba.

group-id

Specifies the group's unique numerical ID (GID) within the system.

Ensure that the command is identical on each node that can run Support for Oracle RAC.
You can create the name service entries in a network name service, such as the Network Information Service (NIS) or NIS+, so that the information is available to the data service clients. You can also create entries in the local /etc files to eliminate dependency on the network name service.
Create the home directory of each potential user in the DBA group that you defined in Step 2.
You are not required to create a home directory for the root user.
For each potential user whose home directory you are creating, type the following command:
```
# mkdir -p user-home
```
user-home

Specifies the full path of the home directory that you are creating.
Add each potential user in the DBA group that you defined in Step 2 to the system.
You are not required to add the root user.

Use the useradd(1M) command to add each user. Adding a user to the system adds an entry for the user to the following files:
- /etc/passwd
- /etc/shadow
```
# useradd -u user-id -g group-name -d user-home \
[ -s user-shell] user-name
```
-u user-id

Specifies the user's unique numerical ID (UID) within the system.

-g group-name

Specifies the name of the user group of which the user is a member. You must specify the DBA group that you defined in Step 2.

-d user-home

Specifies the full path of the user's home directory. You must specify the home directory that you created for the user in Step 3.

-s user-shell

Optionally specifies the full path name of the program that is to be used as the user's shell when the user logs in. If you omit the -s option, the system uses the /bin/sh program by default. If you specify the -s option, user-shell must specify a valid executable file.

user-name

Specifies the user name of the user that you are adding. You must specify the name of a potential user in the DBA group that you defined in Step 2.

Ensure that each user is identical on each node that can run Support for Oracle RAC.
Set the password of each user that you added in Step 4.
Use the passwd(1) command to set the password of each user.
1. Type the following command:
```
# password user-name
```
  user-name
  
  Specifies the user name of the user whose password you are setting. You must specify the name of a user in the DBA group that you added in Step 4.
  
  The passwd command prompts you for the password.
2. In response to the prompt, type the password and press Return.
  The passwd command prompts you to retype the password.
3. In response to the prompt, retype the password and press Return.
Change the ownership of each home directory that you created in Step 3 as follows:
- Owner: the user for which you created the home directory
- Group: the DBA group that you defined in Step 2
For each home directory for which you are changing ownership, type the following command:
```
# chown user-name:group-name user-home
```
user-name

Specifies the user name of the user for whose home directory you are changing ownership. You must specify the name of a user in the DBA group that you added in Step 4.

group-name

Specifies the name of the user group of which the user is a member. You must specify the DBA group that you defined in Step 2.

user-home

Specifies the full path of the user's home directory. You must specify the home directory that you created for the user in Step 3.
Create a subdirectory of the /var/opt directory for each user in the DBA group that you added in Step 4.
For each subdirectory that you are creating, type the following command:
```
# mkdir /var/opt/user-name
```
user-name

Specifies the user name of the user whose subdirectory of the /var/opt directory you are creating. You must specify the name of a user in the DBA group that you added in Step 4.
Change the ownership of each directory that you created in Step 7 as follows:
- Owner: the user for which you created the directory
- Group: the DBA group that you defined in Step 2
For each directory for which you are changing ownership, type the following command:
```
# chown user-name:group-name /var/opt/user-name
```
user-name

Specifies the user name of the user for whose home directory you are changing ownership. You must specify the name of a user in the DBA group that you added in Step 4.

group-name

Specifies the name of the user group of which the user is a member. You must specify the DBA group that you defined in Step 2.

Example 1-1 Creating the DBA Group and the DBA User Accounts

This example shows the sequence of commands for creating the DBA group dba, which is to contain the users root and oracle.

The dba group and the oracle user are created as follows:

The GID of the dba group is 520.
The home directory of the oracle user is /Oracle-home.
The UID of the oracle user is 120.
The oracle user's login shell is the Korn shell.

# groupadd -g 520 dba
# mkdir /Oracle-home
# useradd -u 120 -g dba -d /Oracle-home -s /bin/ksh oracle
# passwd oracle
New Password:oracle
Re-enter new Password:oracle
passwd: password successfully changed for oracle
# chown oracle:dba /Oracle-home
# mkdir /var/opt/oracle
# chown oracle:dba /var/opt/oracle

How to Configure Shared Memory for Oracle RAC Software in the Global Cluster

To enable the Oracle RAC software to run correctly, you must ensure that sufficient shared memory is available on all the cluster nodes. Perform this task on each cluster node.

Become superuser on a cluster node.
Update the /etc/system or /etc/project file with the shared memory configuration information.
Follow the procedure for configuring kernel parameters in your Oracle Real Application Clusters installation guide.
You must configure these parameters on the basis of the resources that are available in the cluster. However, the value of each parameter must be sufficient to enable the Oracle RAC software to create a shared memory segment that conforms to its configuration requirements. For the minimum required value of each parameter, see your Oracle documentation.
The following example shows entries to configure in the /etc/system file.
```
*SHARED MEMORY/ORACLE
set shmsys:shminfo_shmmax=4294967295
```

If you updated the /etc/system file, shut down and reboot each node whose /etc/system file you updated in Step 2.

No reboot is necessary after updating an /etc/project file.

Caution - Before you reboot, you must ensure that any volume manager software that is already installed is completely installed. An incomplete installation of volume manager software causes a panic. If you use VxVM, also ensure that you have installed the software and that the license for the VxVM cluster feature is valid. Otherwise, a panic occurs when you reboot. For information about how to recover from a node panic during installation, see Node Panic During Initialization of Support for Oracle RAC.

For detailed instructions, see Shutting Down and Booting a Single Node in a Cluster in Oracle Solaris Cluster System Administration Guide.

Next Steps

Go to Installing the Support for Oracle RAC Packages.

How to Configure Shared Memory for Oracle RAC Software in a Zone Cluster

To configure shared memory for the Oracle RAC software in a zone cluster, perform the following steps:

Perform the steps to configure the shared memory in a global cluster.
For detailed instructions on configuring the shared memory in a global cluster, see How to Configure Shared Memory for Oracle RAC Software in the Global Cluster.
Perform the similar steps in each zone cluster.
Note - These steps do not affect the actual shared memory control for the zone cluster. You perform these steps to help the Oracle dbca utility enable you to set the database memory allocation. If the Oracle dbca utility is not used for Oracle RAC database creation, you can skip these steps in the zone cluster.
Perform the following steps, if you want to limit the memory used for the zone cluster.
1. Become superuser on the global cluster node that hosts the zone cluster.
2. Configure the capped-memory property attributes physical, swap, and locked by using the clzonecluster command.
```
#clzonecluster configure zcname
clzonecluster:zcname> add capped-memory
clzonecluster:cz1-2n:capped-memory> set physical=memsize
clzonecluster:cz1-2n:capped-memory> set swap=memsize
clzonecluster:cz1-2n:capped-memory> set locked=memsize
clzonecluster:cz1-2n:capped-memory> end
clzonecluster:cz1-2n>commit
```
  Physical=memsize
  
  Specifies the physical memory size.
  
  swap=memsize
  
  Specifies the swap memory size.
  
  locked=memsize
  
  Specifies the limit of the shared memory segment size that the Oracle RAC database processes can request to lock in memory.
  
  Note - In addition to the locked attribute of the capped-memory property, you can use the max-shm-memory property to directly configure the limit of the shared memory segment in a zone cluster. See also the Solaris man page zonecfg(1M).
3. Reboot the zone cluster.
```
#clzonecluster reboot zcname
```
  Note - You can perform the step to configure the capped-memory property attributes as part of the zone cluster creation. If you configure the capped-memory property attributes as part of the zone cluster creation, the memory-related properties immediately take effect after the first zone cluster boot. See How to Create a Zone Cluster in Oracle Solaris Cluster Software Installation Guide.

How to Set the Necessary Privileges for Oracle RAC Software in a Zone Cluster

You should set the necessary privileges to enable Oracle RAC to run in a zone cluster configuration. You can use the clzonecluster command to include the necessary privileges in a zone cluster configuration by setting the limitpriv property. Perform the following steps to set the necessary privileges in a zone cluster to run Oracle RAC.

Become superuser on the global cluster node that hosts the zone cluster.

Configure the limitpriv property by using the clzonecluster command.

# clzonecluster configure zcname
clzonecluster:zcname>set limitpriv ="default,proc_priocntl,proc_clock_highres"
clzonecluster:zcname>commit

Beginning with Oracle RAC version 11g release 2, prevent Oracle Clusterware time synchronization from running in active mode.
1. Log in to the zone-cluster node as root.
2. Create an empty /etc/inet/ntp.conf file.
```
# touch /etc/inet/ntp.conf
```

How to Configure the Logical Hostname Resources or Virtual IP Addresses for Oracle RAC Software in a Zone Cluster

To support logical hostname resources for an Oracle RAC 9i configuration and Oracle Clusterware virtual IP resources in Oracle RAC 10g, 11g, and 12c configurations in zone clusters, you should configure the failover-capable hostnames or IP addresses used by those resources in a given zone cluster using the clzonecluster command.

Perform the following steps to configure the virtual IP addresses in a zone cluster configuration for Oracle RAC 10g, 11g, or 12c.

Become superuser on the global cluster node that hosts the zone cluster.

Configure the virtual IP addresses using the clzonecluster command.

# clzonecluster configure zcname
clzonecluster:zcname>add net
clzonecluster:zcname:net>set address=racnode1-vip
clzonecluster:zcname:net>end
clzonecluster:zcname>add net
clzonecluster:zcname:net>set address=racnode2-vip
clzonecluster:zcname:net>end
clzonecluster:zcname>commit

Reboot the zone cluster.
```
# clzonecluster reboot zcname
```

Skip Navigation Links
Exit Print View
	Oracle Solaris Cluster Data Service for Oracle Real Application Clusters Guide Oracle Solaris Cluster